Scalable toolkit for efficient model reinforcement
Python · active 2025-06-12 → 2026-08-10 (UTC)
Activity over time
Daily event counts in the loaded window
Line chart, 425 days from 2025-06-12 to 2026-08-10. Pushes: 5,428 total, peak 48 in a day. Pull requests: 1,889 total, peak 28 in a day. Issues: 854 total, peak 36 in a day. Comments: 4,436 total, peak 108 in a day. Stars: 519 total, peak 45 in a day.
- Pushes
- Pull requests
- Issues
- Comments
- Stars
Stars, PRs, issues and forks are under-captured in the later part of this window. GH Archive progressively stopped capturing non-push events during 2026 — −95% or worse by the end of the window. Every series here except Pushes fades for that reason, so a decline above reflects the archive, not this repository. Pushes stay reliable throughout, so read them, and the contributor counts derived from them, as the real signal. Data health has the measurements.
Top contributors
Pushes, PRs, issues, reviews and comments — stars and forks excluded, so this is contribution rather than popularity
| Contributor | Contributions | Pushes | PRs | Comments |
|---|---|---|---|---|
| terrykong | 2.9K | 867 | 425 | 737 |
| coderabbitai[bot] | 1.1K | 1 | 2 | 700 |
| yuki-97 | 926 | 393 | 144 | 212 |
| github-actions[bot] | 853 | 0 | 136 | 665 |
| copy-pr-bot[bot] | 784 | 694 | 0 | 90 |
| chtruong814 | 576 | 242 | 203 | 38 |
| ashors1 | 541 | 270 | 38 | 117 |
| yfw | 480 | 226 | 65 | 73 |
| guyueh1 | 430 | 53 | 104 | 158 |
| RayenTian | 422 | 191 | 92 | 72 |
| parthchadha | 300 | 65 | 30 | 60 |
| ZhiyuLi-Nvidia | 273 | 126 | 22 | 72 |
| yuki-666 | 224 | 82 | 10 | 65 |
| youngeunkwon0405 | 218 | 88 | 67 | 37 |
| SahilJain314 | 215 | 67 | 24 | 53 |
| ffrujeri | 199 | 103 | 14 | 44 |
| seonjinn | 197 | 182 | 4 | 10 |
| bxyu-nvidia | 194 | 131 | 13 | 19 |
| snowmanwwg | 184 | 15 | 7 | 36 |
| jveronvialard | 176 | 31 | 3 | 74 |
Recent activity
Latest issues, pull requests and releases
- Pull request#3543nvegesna-netizen2026-08-07 19:34
- Issue comment#2962yfw2026-08-07 18:59fix: sanitize non-finite logprobs in vllm async worker
- Issue comment#3538svcnvidia-nemo-ci2026-08-07 18:40ci: AUT-1397 support non-root local container tests
- Issue comment#3502slikhite-12026-08-07 18:16feat(glm): add GLM 5.1 and 5.2 support
- Issue comment#3391jthomson042026-08-07 18:10feat(dynamo): add managed Slurm generation with W&B telemetry
- Issue comment#1126guyueh12026-08-07 17:28perf: Support mcore virtual pipeline parallel
- Issue comment#2368jthomson042026-08-07 17:13fix(async-grpo): fail fast on generation worker errors
- Issue comment#3523copy-pr-bot[bot]2026-08-06 15:43fix(megatron): plumb moe_router_enable_expert_bias from megatron_cfg
- Issue comment#2625terrykong2026-07-30 06:02[Single Controller / Async RL] cleanup tracking issue
- Pull request#3405tdene2026-07-29 11:32
- Pull request#3389github-actions[bot]2026-07-28 22:19
- Pull request#2420joyang-nv2026-07-22 03:14
- Pull request#3295seonjinn2026-07-21 22:10
- Issue comment#2983github-actions[bot]2026-07-19 00:20feat(modelopt): support real NVFP4 QAT rollout for MoE and Mamba
- Issue comment#3273chtruong8142026-07-17 23:05cp: `preserve runtime tools after arm64 Apptainer build (3253)` into `r0.7.0`
- Issue comment#3227github-actions[bot]2026-07-16 23:43feat(nemotron-omni): add Megatron Nano image GRPO support
- Pull request#3236svcnvidia-nemo-ci2026-07-16 08:42
- Pull request#2681ZhiyuLi-Nvidia2026-07-15 23:14
- Issue comment#3207parthmannan2026-07-15 18:13ci: Nemotron3-Super Perf Recipe
- Pull request#3202kajalj222026-07-14 23:42
- Issue comment#3201copy-pr-bot[bot]2026-07-14 23:12fix: honor NeMo Gym's mask_sample in the Gym rollout path
- Pull request#3186github-actions[bot]2026-07-14 00:43
- Issue comment#3178yuki-972026-07-13 13:04fix: ensure unique TOPO_RANK per node
- Issue comment#3131RayenTian2026-07-13 05:48feat: add OpenR1-Math-220k response dataset
- Issue#1128svcnvidia-nemo-ci2026-07-10 19:30megatron.core.dist_checkpointing.core.CheckpointingException: Invalid access pattern: 36 ShardedObject are missing
Totals cover only the window loaded into ClickHouse and count events, not GitHub's lifetime totals — 519 stars here means stars gained during the window, not the repo's star count.